NASA Is Paying to Survey the Moon. Nobody Knows What Is Actually There.

The U.S. government just agreed to buy something nobody has extracted yet. Three liters of helium-3, delivery due April 2029, from a company that has never shipped a single gram of lunar material. The contract with the Department of Energy Isotope Program is real, the amount is undisclosed, and it is the first American purchase order for a resource that does not yet exist on Earth.

That purchase order is the demand signal behind NASA's $6.9 million award to Interlune of Seattle, announced this month, to develop and test ISRU prospecting hardware: a regolith collector, particle sorter, and gas extraction system built around a mass spectrometer called MSOLO. The goal is to make future lunar missions less dependent on supplies shipped from Earth. The contract runs for a year and a half.

It is a real contract for a real amount of money. MSOLO is real hardware that has flown before. What Interlune is selling — and what NASA and the DoE are both funding — is the prospector's legwork. The company emerged from stealth in early 2025 with $18 million in seed funding, a full-scale excavator prototype built in partnership with industrial equipment manufacturer Vermeer, and a four-step harvesting system designed to process 100 metric tons of regolith per hour. To deliver three liters of helium-3, Interlune calculates it needs to process enough regolith to fill a large backyard swimming pool.

Helium-3 is scarce on Earth but theoretically abundant on the Moon. It has uses in national security screening, quantum computing cooling systems, and medical imaging. It is also a candidate fuel for future fusion reactors, a market that does not yet exist at commercial scale. "A sizable enough quantity to require extracting and separating it from the regolith while on the Moon instead of transporting the regolith back to Earth for processing here," Interlune CEO Rob Meyerson said in the company's announcement.

The survey that will determine whether any of this math holds is the part that does not exist yet. In March 2025, MSOLO rode aboard Intuitive Machines' IM-2 mission to the lunar south pole aboard a Nova-C lander named Athena. Athena landed on its side inside a crater, 400 meters from its intended site, and depleted its batteries before completing most of its science objectives. According to NASA's own post-mission summary, MSOLO "detected elements likely due to the gases emitted from the lander's propulsion system." It did not report confirmed concentrations of water ice, helium-3, or any other volatiles at the south pole. The instrument worked. The data is not there.

That gap is the specific hardware problem Interlune's NASA contract is meant to close. MSOLO's IM-2 flight demonstrated the spectrometer can survive a lunar landing, even a botched one — useful engineering data, but not a resource assessment. The South Pole's permanently shadowed craters are candidates for water ice accumulation, but their distribution, depth, and purity are still modeled, not measured. Interlune has not published a site selection.

The commercial case for lunar mining rests on a handful of prior data points: orbital spectroscopy suggesting water ice concentrations at the poles, Apollo-era samples indicating helium-3 deposition from the solar wind, and the theoretical physics of how volatiles accumulate in permanently shadowed regions. These are real foundations. They are also projections, not measurements. Whether the regolith at the right locations contains enough helium-3 — or water ice, which would support in-situ propellant production for the Artemis program — to make the operation economically coherent is the question the $6.9 million is buying the hardware to answer.

The historical parallel that fits is not the California Gold Rush. It is the early seismic survey business. In the 1920s and 1930s, before a single exploratory well was drilled, companies like Geophysical Services Inc. — later Texas Instruments — sold the maps that let oil companies decide where to spend drilling capital. The surveyors were not the extractors. They were the people who made the extractors' bets rational. Interlune is positioning itself the same way: as the measurement layer that determines where actual mining operations should go.

Whether that position holds depends entirely on what the first surveys find. If Interlune's instruments find rich deposits, the company has a defensible position as the sector's geological surveyor. If the concentrations are too diffuse to justify the extraction cost, the entire commercial architecture — the DoE purchase, the Vermeer partnership, the Artemis propellant depots — needs to be rethought.

The DoE's helium-3 purchase is not a guarantee the deposits are there. It is a bet that the survey will answer the question in time to meet a delivery deadline in 2029.